Direct solar radiation prediction based on soft-computing algorithms including novel predictive atmospheric variables

Abstract

In this paper we tackle a problem of solar radiation prediction with Soft-Computing Techniques. We introduce new atmospheric input variables in the problem, which help to obtain an accurate prediction of solar radiation. We test the performance of two state-of-the art algorithms: Extreme Learning Machines and Support Vector regression algorithms, in a real problem of solar radiation prediction in Murcia, Spain, where we have obtained excellent results with the proposed techniques. © 2013 Springer-Verlag.